It seems to be a slightly different statement: "We're not going to enable Mac OS 10 to run on machines other than Apple Macintoshes". The other quotes I've seen have said "allow". The rest of the exchange was garbled, can someone make out exactly what was said?
Question: "So you're not [considering?] [something] a hack [unintelligable]" Answer: "*shrug* Who knows what the future holds, what people won't do, but that's how they're [working on it?]."
If you're worried about someone other than Intel/AMD coming along and delivering the Next Big Thing(tm), well, the universe of Mac code is going to be a heck of a lot more portable after next year, won't it?
This is the important bit.
Apple tried to make this happen in 1997, when they announced Rhapsody and wanted everyone to code for Yellow Box.
They got handed their heads by Adobe and the rest, back then.
there's no reason to do this now, other than all the reasons that make it not only practical, but the perfect time
But it's the right time because they can get away with it. Not because they suddenly went "Oh My God, The Power PC Stinks". It's not stinking any worse now than it has at almost any other time in the past five years. It's actually looking better in the medium term than it has for most of that time.
I'm convinced the loss of Carbon-chained developers will be miniscule. Cocoa is just so much easier to develop new apps in that most of the major Carbon users left are the big development houses
Plus absolutely anyone who's doing cross-platform work, since there's no OpenStep any more, GNUStep has languished, and there's no Objective-C.NET. That's most of the open source community, to start with, but even Apple's iTunes is Carbon, as far as I can tell, because that's what it takes to be portable.
That's undoubtedly why they've made sure they can build Carbon universal binaries... but they have to be Mach-O, not CEF. Which means it's not as big a problem as it could be... but it's still easier under Cocoa.
It would be a real smart move for Apple, though, to give GNUstep a thumbs up and to re-release OpenStep for Windows... so people can write apps that work superbly on OS X and decently on Linux and Windows.
They tell us not to be "zeolot" or "fanboy", why the heck OS X matters than?
One doesn't need to be zealot or fanboy to prefer an OS that doesn't suck. Without OS X, your choice is a Windows with 30 years of bad design choices baked into it (not to mention it's increasingly crippled by copy protection and digital rights management) or free UNIX with no commercial desktop apps. Or something even more marginal like BeOS or Yellowtab.
Apparently you don't fuck up the air flow by putting drives there:
When I tested the Swift, the only concern I had was the effect the drives "invading" the CPU bay might have on the temperature control system of the G5. The fan that blows air over the CPUs now sucks air over the three drives. How will those drives affect air flow? Will drives heat up the air being blown over the CPUs? I used to ThermographX to monitor the 7 temperature probes in the G5. The temperature readings were not significantly affected by the presence of the extra drives. -- Five Drives in a G5
FIREWIRE.. portable, non-ugly, and don't screw with the air flow
Higher latency, more cables to deal with, and makes the system as a whole less portable.
Though I wouldn't have bothered with the cage: he could just have screwed the drives directly to the grille. Less interference with the air flow and the cabling's easier.
Huh, I thought things were more concurrent than that.
Cloning was bleeding apple dry because the clones were still basically Macs as far as the rest of the world was concerned, so it didn't expand the user-base any... it just cannibalized Apple's sales.
NeXTSTeP/Rhapsody/OS X on "generic Intel" would cut into Mac sales some, but it would be more likely to expend the user base because people wouldn't have to buy a custom computer just to run it.
But, that's Steve's call. I can see good arguments either way.
It's like slapping bald snow tires on the back that are too big and rub against the fenders, then.
Unless you know something about those drives that I don't, they're not "bald".
As for the rest: if you have a Civic and you need to drive through snow, well, you gotta do what you gotta do. Just don't turn too tightly and you'll have plenty of clearance.
Re:This is my experience with FreeBSD
on
FreeBSD 5.4 Review
·
· Score: 4, Insightful
One employee lost a whole project due to the OS corrupting the filesystem.... A few days later I was looking for another job
So you were fired because you hadn't been making backups, right?
I would buy one now, or wait until 2007. Maybe late 2007.
You're not likely to get much better of a deal from Apple between now and next June, and it's going to take them a while to get up to speed, so if you're in the market for a Mac now you'll get the most use out of any Mac you'll get between now and then: based on past transitions Apple will have OS upgrades for you through 2010... and they'll track them feature-for-feature because that's easy, so apart from games there won't be any incentive for ISVs to *not* produce fat binary versions that long.
In 2007 you'll be able to get used Powermacs for a real good price as the x86 Macs come in, as well as the x86 Macs themselves.
So you might want to do something like getting a Mac mini now, and plan on upgrading in 2007.
I am all for supporting Intel, but Mac OS X is platform-neutral by design, which is a huge advantage for our platform and shouldn't be casually thrown away.
I suspect one reason they're doing this is to make it *really* platform neutral, to flush out the last of the heritage of the old Mac OS.
Because if it was really platform-neutral like NeXTstep was, this wouldn't be a big deal. But it's not. It's not just Classic, either... Carbon is leading two different lives, and only one of these lives has a future. There's CEF-format executables that are PPC only and run on Mac OS 9 as well as Mac OS X, and there's Mach-O executables that are OSX-only but can be turned into fat... I mean universal... binaries.
That'll make it a real challenge, since the designers of Java made an effort to make it difficult to write malicious code in the first place.
Actually, that's not really the case... not for the kind of "malicious code" that they're talking about here. They're not talking about "getting out of the sandbox", they're talking about "hiding information in the output". It's actually a lot easier to hide this kind of "malicious code" in an object-oriented language because you can play games with the namespace.
It's unlikely that any techniques that really pass inspection will be C-specific. "Obfuscated coding contest tricks" won't help, because the code has to look benign, and weird comments and variable names and odd spacing won't help.
Actually, having the trashcan move around is really annoying, but it's nothing to do with "Fitt's Law" (which, as I said, most Mac zealots don't even understand)... it's just that hitting a moving target to delete a file is like playing a videogame and I suck at videogames.
Luckily you can put a delete button on the finder toolbar and never care about the "trashcan metaphor" ever again.
It will take a re-compilation of the OS to execute on x86 processors.
Yeh, but that's not a big deal any more. Back in the '80s switching a bunch of high performance graphics software from one CPU to another meant rewriting a bunch of highly optimized assembly code and tweaking it for the new platform. But now it's pretty much a matter of changing a few compiler options... most of the really hairy work is in the video drivers and graphics card now, and OpenGL means most of THAT is portable.
Why Apple switched now is that Apple was able to switch now. They've just moved a bunch more graphics code to the GPU with QE2d, so there's less dependence on getting the absolute last bit of performance from the CPU that there's ever been. And on the marketing side they've finally put a stake through the heart of OS 9 by dropping the last OS 9 bootable Mac. Switching the CPU will let them cut off the head of OS 9 and stuff its mouth with garlic for good measure. There's nothing holding them back from completing the transition they started by bringing Steve Jobs back from NeXT with an actual operating system instead of a bunch of really great graphics libraries tied together with spit and baling wire that they called Mac OS.
I suspect that Leopard, even on the PPC, won't run on any Mac that can't support Quartz Extreme. That way it'll run really fast even if the codebase isn't quite as portable as they hoped.
I don't like the processor change, but if that's what it takes to kill the classic Mac OS for good I guess it's worth it.
Apple's a company that makes great software and pretty good hardware that makes its money by using the software to sell the hardware, so they THINK they're still a hardware company.
On the other hand, Microsoft's a company that makes a little bit of great hardware and some good applications, but is mostly a holding company for some legacy intellectual property that's very valuable. They make most of their money by licensing that IP... so they THINK they're a software company.
Kind of like SCO was, back when they'd given up really improving UnixWhatever but before their IP had lost its value. They were smart enough to get out of that business and let Caldera take it on, and we all know what happened when THEY realised the IP they'd bought wasn't worth anything...
They're also still making itaniums and xscales [...]
Yeh, they've got a whole bunch of processors that don't have anything to do with personal computers. I thought we were talking about personal computers, though. I mean, you know, you already mentioned that IBM and Freescale have bunches of other processors that don't have anything to do with personal computers, but for them that was supposed to be a bad thing.
Your argument boils down to "the G5 would be faster than intel's previous top-of-the line if apple would just rearchitect sometime in the future."
IDGI. The text you were replying to wasn't even talking about the G5.
People aren't getting it. I don't know anyone, *anyone* who will buy an OS upgrade.
I bought Windows 2000. I didn't buy XP because it was a downgrade from 2000, but 2000 was an improvement over 9x and NT4 (heck, NT3.51 was an improvement over NT4).
I also bought Jaguar and Panther and as soon as I get the time set aside to actually do the upgrade I'll be buying Tiger. But then newer versions of Mac OS X are more stable and run faster, and even with a short release cycle they manage to include useful new capabilities.
So many people went to Windows XP because even those who used Windows 2000 saw a lot of good benefits in it.
I'm still using 2000 when I use Windows, and I have no intention to jump to XP and its "time bomb" kernel no matter how often Microsoft says it's "obsolete". I've used XP at work, and while it's faster in some places it's slower in others (especially after SP2), and an operating system is too important a component of an OS to put up with strong copy protection inside it.
[By the way, if Apple puts strong copy protection in Leopard to keep people from using it on clones, I guess I'll be sticking with Panther or Tiger for the forseeable future.]
Windows XP was a win for most people because they were upgrading from 98 or Me, and they upgraded because Microsoft had screwed up Me so badly, and there was no further upgrade path from 9x.
How about a shorter list... features that aren't going to end up in XP anyway, and features that actually have value.
Avalon: a new user interface subsystem and API based on XML,.NET, and vector graphics.
Which will also be available for XP. Scratch one.
WinFX: a new API replacing the current Win32 API (there's of course still Win32 + Win64).NET framework 2.0 (the foundation for Longhorn)
That's two (or is it three) new APIs. New APIs by themselves have negative value. What can you do with them... that's the important bit.
Lower user privileges (IE 7 will run in these on Longhorn)
Fixing the wrong problem. The only reason to run IE in some kind of sandbox is because of its broken active content model. Instead they should fix IE by backing that out and split off a local HTML scripting environment (like Dashboard, but without the stupid UI), and making IE into a normal browser... a purely web aplication that has no ability to download applets and automatically* run them with full local rights.
Included compiler (msbuild)
Ah, finally catching on to what every UNIX vendor figured out by the early '90s. Even SCO ecentually "got" that. Of course you can download SFU and get an included compiler AND a decent shell, RIGHT NOW... so this is also available for XP and Windows 2000.
New document format competitive to PDF
Something else with negative value.
An application deployment engine ("ClickOnce")
Sounds Linspired. I hope they've actually thought about security this time.
New desktop search capabillities
Already available for XP, not a Longhorn feature.
Improved security through lower privileged accounts and services
You're repeating yourself.
* popping up a routine dialog box that people are used to clicking OK on is hardly better than running it with no warning.
Do you agree the G5 is more efficient that the G4?
Not at all. The G5 is more like the P4... oh, it's a better implementation of the long-pipeline super-brainiac design than the P4, but it's still a high-clock low-efficiency approach. That's why the whole idea of a "G5 Powerbook" is just crazy.
If it wasn't for the low speed bus, I wouldn't be at all surprised to see the G4 beating the G5 handily, clock-for-clock.
Slashdot Mirror
It seems to be a slightly different statement: "We're not going to enable Mac OS 10 to run on machines other than Apple Macintoshes". The other quotes I've seen have said "allow". The rest of the exchange was garbled, can someone make out exactly what was said?
Question: "So you're not [considering?] [something] a hack [unintelligable]"
Answer: "*shrug* Who knows what the future holds, what people won't do, but that's how they're [working on it?]."
If you're worried about someone other than Intel/AMD coming along and delivering the Next Big Thing(tm), well, the universe of Mac code is going to be a heck of a lot more portable after next year, won't it?
This is the important bit.
Apple tried to make this happen in 1997, when they announced Rhapsody and wanted everyone to code for Yellow Box.
They got handed their heads by Adobe and the rest, back then.
there's no reason to do this now, other than all the reasons that make it not only practical, but the perfect time
But it's the right time because they can get away with it. Not because they suddenly went "Oh My God, The Power PC Stinks". It's not stinking any worse now than it has at almost any other time in the past five years. It's actually looking better in the medium term than it has for most of that time.
I'm convinced the loss of Carbon-chained developers will be miniscule. Cocoa is just so much easier to develop new apps in that most of the major Carbon users left are the big development houses
Plus absolutely anyone who's doing cross-platform work, since there's no OpenStep any more, GNUStep has languished, and there's no Objective-C.NET. That's most of the open source community, to start with, but even Apple's iTunes is Carbon, as far as I can tell, because that's what it takes to be portable.
That's undoubtedly why they've made sure they can build Carbon universal binaries... but they have to be Mach-O, not CEF. Which means it's not as big a problem as it could be... but it's still easier under Cocoa.
It would be a real smart move for Apple, though, to give GNUstep a thumbs up and to re-release OpenStep for Windows... so people can write apps that work superbly on OS X and decently on Linux and Windows.
They tell us not to be "zeolot" or "fanboy", why the heck OS X matters than?
One doesn't need to be zealot or fanboy to prefer an OS that doesn't suck. Without OS X, your choice is a Windows with 30 years of bad design choices baked into it (not to mention it's increasingly crippled by copy protection and digital rights management) or free UNIX with no commercial desktop apps. Or something even more marginal like BeOS or Yellowtab.
FIREWIRE .. portable, non-ugly, and don't screw with the air flow
Higher latency, more cables to deal with, and makes the system as a whole less portable.
Though I wouldn't have bothered with the cage: he could just have screwed the drives directly to the grille. Less interference with the air flow and the cabling's easier.
10-4 good buddy, the Powermac G5 is simply overcompensation for the noise of the last generation Powermac G4s.
Huh, I thought things were more concurrent than that.
Cloning was bleeding apple dry because the clones were still basically Macs as far as the rest of the world was concerned, so it didn't expand the user-base any... it just cannibalized Apple's sales.
NeXTSTeP/Rhapsody/OS X on "generic Intel" would cut into Mac sales some, but it would be more likely to expend the user base because people wouldn't have to buy a custom computer just to run it.
But, that's Steve's call. I can see good arguments either way.
It's like slapping bald snow tires on the back that are too big and rub against the fenders, then.
Unless you know something about those drives that I don't, they're not "bald".
As for the rest: if you have a Civic and you need to drive through snow, well, you gotta do what you gotta do. Just don't turn too tightly and you'll have plenty of clearance.
One employee lost a whole project due to the OS corrupting the filesystem. ... A few days later I was looking for another job
So you were fired because you hadn't been making backups, right?
Some people buy Macs because they want to actually use them, not because they want to admire the case from a distance.
About the only thing I could compare this to is slapping narrow high profile snow tires on the back of a civic
Hardly. It's more like slapping snow tires on a Civic because you're going to drive through snow.
I would buy one now, or wait until 2007. Maybe late 2007.
You're not likely to get much better of a deal from Apple between now and next June, and it's going to take them a while to get up to speed, so if you're in the market for a Mac now you'll get the most use out of any Mac you'll get between now and then: based on past transitions Apple will have OS upgrades for you through 2010... and they'll track them feature-for-feature because that's easy, so apart from games there won't be any incentive for ISVs to *not* produce fat binary versions that long.
In 2007 you'll be able to get used Powermacs for a real good price as the x86 Macs come in, as well as the x86 Macs themselves.
So you might want to do something like getting a Mac mini now, and plan on upgrading in 2007.
I am all for supporting Intel, but Mac OS X is platform-neutral by design, which is a huge advantage for our platform and shouldn't be casually thrown away.
I suspect one reason they're doing this is to make it *really* platform neutral, to flush out the last of the heritage of the old Mac OS.
Because if it was really platform-neutral like NeXTstep was, this wouldn't be a big deal. But it's not. It's not just Classic, either... Carbon is leading two different lives, and only one of these lives has a future. There's CEF-format executables that are PPC only and run on Mac OS 9 as well as Mac OS X, and there's Mach-O executables that are OSX-only but can be turned into fat... I mean universal... binaries.
Ah, so they're not just crippled versions of the 970, they may be a separate parallel development path.
That'll make it a real challenge, since the designers of Java made an effort to make it difficult to write malicious code in the first place.
Actually, that's not really the case... not for the kind of "malicious code" that they're talking about here. They're not talking about "getting out of the sandbox", they're talking about "hiding information in the output". It's actually a lot easier to hide this kind of "malicious code" in an object-oriented language because you can play games with the namespace.
It's unlikely that any techniques that really pass inspection will be C-specific. "Obfuscated coding contest tricks" won't help, because the code has to look benign, and weird comments and variable names and odd spacing won't help.
Then the prize is near-beer.
Actually, having the trashcan move around is really annoying, but it's nothing to do with "Fitt's Law" (which, as I said, most Mac zealots don't even understand)... it's just that hitting a moving target to delete a file is like playing a videogame and I suck at videogames.
Luckily you can put a delete button on the finder toolbar and never care about the "trashcan metaphor" ever again.
It will take a re-compilation of the OS to execute on x86 processors.
Yeh, but that's not a big deal any more. Back in the '80s switching a bunch of high performance graphics software from one CPU to another meant rewriting a bunch of highly optimized assembly code and tweaking it for the new platform. But now it's pretty much a matter of changing a few compiler options... most of the really hairy work is in the video drivers and graphics card now, and OpenGL means most of THAT is portable.
Why Apple switched now is that Apple was able to switch now. They've just moved a bunch more graphics code to the GPU with QE2d, so there's less dependence on getting the absolute last bit of performance from the CPU that there's ever been. And on the marketing side they've finally put a stake through the heart of OS 9 by dropping the last OS 9 bootable Mac. Switching the CPU will let them cut off the head of OS 9 and stuff its mouth with garlic for good measure. There's nothing holding them back from completing the transition they started by bringing Steve Jobs back from NeXT with an actual operating system instead of a bunch of really great graphics libraries tied together with spit and baling wire that they called Mac OS.
I suspect that Leopard, even on the PPC, won't run on any Mac that can't support Quartz Extreme. That way it'll run really fast even if the codebase isn't quite as portable as they hoped.
I don't like the processor change, but if that's what it takes to kill the classic Mac OS for good I guess it's worth it.
Apple's a company that makes great software and pretty good hardware that makes its money by using the software to sell the hardware, so they THINK they're still a hardware company.
On the other hand, Microsoft's a company that makes a little bit of great hardware and some good applications, but is mostly a holding company for some legacy intellectual property that's very valuable. They make most of their money by licensing that IP... so they THINK they're a software company.
Kind of like SCO was, back when they'd given up really improving UnixWhatever but before their IP had lost its value. They were smart enough to get out of that business and let Caldera take it on, and we all know what happened when THEY realised the IP they'd bought wasn't worth anything...
They're also still making itaniums and xscales [...]
Yeh, they've got a whole bunch of processors that don't have anything to do with personal computers. I thought we were talking about personal computers, though. I mean, you know, you already mentioned that IBM and Freescale have bunches of other processors that don't have anything to do with personal computers, but for them that was supposed to be a bad thing.
Your argument boils down to "the G5 would be faster than intel's previous top-of-the line if apple would just rearchitect sometime in the future."
IDGI. The text you were replying to wasn't even talking about the G5.
People aren't getting it. I don't know anyone, *anyone* who will buy an OS upgrade.
I bought Windows 2000. I didn't buy XP because it was a downgrade from 2000, but 2000 was an improvement over 9x and NT4 (heck, NT3.51 was an improvement over NT4).
I also bought Jaguar and Panther and as soon as I get the time set aside to actually do the upgrade I'll be buying Tiger. But then newer versions of Mac OS X are more stable and run faster, and even with a short release cycle they manage to include useful new capabilities.
So many people went to Windows XP because even those who used Windows 2000 saw a lot of good benefits in it.
I'm still using 2000 when I use Windows, and I have no intention to jump to XP and its "time bomb" kernel no matter how often Microsoft says it's "obsolete". I've used XP at work, and while it's faster in some places it's slower in others (especially after SP2), and an operating system is too important a component of an OS to put up with strong copy protection inside it.
[By the way, if Apple puts strong copy protection in Leopard to keep people from using it on clones, I guess I'll be sticking with Panther or Tiger for the forseeable future.]
Windows XP was a win for most people because they were upgrading from 98 or Me, and they upgraded because Microsoft had screwed up Me so badly, and there was no further upgrade path from 9x.
How about a shorter list... features that aren't going to end up in XP anyway, and features that actually have value.
.NET, and vector graphics.
.NET framework 2.0 (the foundation for Longhorn)
Avalon: a new user interface subsystem and API based on XML,
Which will also be available for XP. Scratch one.
WinFX: a new API replacing the current Win32 API (there's of course still Win32 + Win64)
That's two (or is it three) new APIs. New APIs by themselves have negative value. What can you do with them... that's the important bit.
Lower user privileges (IE 7 will run in these on Longhorn)
Fixing the wrong problem. The only reason to run IE in some kind of sandbox is because of its broken active content model. Instead they should fix IE by backing that out and split off a local HTML scripting environment (like Dashboard, but without the stupid UI), and making IE into a normal browser... a purely web aplication that has no ability to download applets and automatically* run them with full local rights.
Included compiler (msbuild)
Ah, finally catching on to what every UNIX vendor figured out by the early '90s. Even SCO ecentually "got" that. Of course you can download SFU and get an included compiler AND a decent shell, RIGHT NOW... so this is also available for XP and Windows 2000.
New document format competitive to PDF
Something else with negative value.
An application deployment engine ("ClickOnce")
Sounds Linspired. I hope they've actually thought about security this time.
New desktop search capabillities
Already available for XP, not a Longhorn feature.
Improved security through lower privileged accounts and services
You're repeating yourself.
* popping up a routine dialog box that people are used to clicking OK on is hardly better than running it with no warning.
Rhapsody on Intel was Amelio's strategy, and it went out the window when SJ stopped the clones.
I thought that was just a continuation of his NeXT on Intel plans. Or was that also someone else behind the scenes?
Do you agree the G5 is more efficient that the G4?
Not at all. The G5 is more like the P4... oh, it's a better implementation of the long-pipeline super-brainiac design than the P4, but it's still a high-clock low-efficiency approach. That's why the whole idea of a "G5 Powerbook" is just crazy.
If it wasn't for the low speed bus, I wouldn't be at all surprised to see the G4 beating the G5 handily, clock-for-clock.